Efforts to identify new therapeutic approaches to Ph(−) myeloproliferative disease have been bolstered by the observations of constitutive activation of the Jak-Stat signaling pathway in MPD patients. In particular, a single valine to phenyalanine mutation at residue 617 (Jak2-V617F) in Jak2 has been observed in the majority of PV (95%), ET (50-60%) and PMF (50-60%) patients (Table 2, Kralovics et al. (2005); Baxter et al. (2005); Tefferi et al. (2005)). The V617F mutation resides in the region of the Jak2 gene encoding the pseudokinase domain which is thought to function as an autoinhibitory domain to regulate Jak2 tyrosine kinase activity. Mutations in exon 12 of Jak2 which also result in constitutive Jak2 kinase activity are also observed with a lesser frequency (<5% in PV and ET) and are mutually exclusive with Jak2V617F lesions (Pardanari et al.
(2007); Scott et al. (2007)). Jak2 is a member of a non-receptor tyrosine kinase family that also includes Jak1, Tyk2 and Jak3 and function as mediators of cytokine receptor signaling (for review see Murray (2007)). Upon cytokine binding to its cognate receptor, receptor-bound Jak family members are activated and phosphorylate a STAT, a latent transcription factor, which upon Jak-mediated phosphorylation undergoes dimerization and translocation to the nucleus to regulate gene expression. Genetic and biochemical studies have established distinct combinations of engagement of a Jak family member with an individual cytokine receptor. For instance, erythropoietin (EPO), thrombopoietin (TPO) and granuolocyte colony stimulating factor (GM-CSF) receptor engagement results in the predominant activation of Jak2 to mediate downstream signaling. Consistent with the patho-physiology of MPD associated with the Jak2-V617F mutation, these cytokines promote the differentiation and expansion of the cell types underlying PV, ET and PMF, respectively. Unlike other genetic activation events, expression of Jak2-V617F is not sufficient to promote transformation in cell based model systems and has been shown to require the co-expression of the type I cytokine receptors, highlighting an important functional co-dependence of the Jak-cytokine receptor interaction (Lu et al. (2005)). Interestingly, activating mutations in the TPO receptor (MPL, tryptophan to leucine substitution at residue 515) have been identified in MPD patients afflicted with Jak2-V617F negative PMF and ET (5% and 1%, respectively) resulting in constitutive Jak2-Stat activation (Pikman et al. (2006)). These observations indicate that Jak-Stat signal transduction can be activated through mutation in MPD at multiple points in the pathway in a mutually exclusive manner and suggest the possible presence of additional pathway mutations in Jak2-V617F and MPL-W515L negative MPD.
Important validation of Jak2 signaling as a driver of Ph(−) MPD emerged from rodent models where Jak2-V617F mutant signaling was reconstituted in the hematopoietic stem cell compartment. Several laboratories demonstrated that viral transduction of Jak2-V617F into mouse bone marrow and subsequent re-implantation into recipient mice reconstituted several aspects human MPD (Wernig et al. (2006); Lacout et al. (2006); Bumm et al. (2006); Zaeleskas et al. (2006)). These features included elevated hematocrit, splenomegaly from extramedullary hematopoiesis, granulocytosis and bone marrow fibrosis all which are also manifested in polycythemia vera. Interestingly, unlike the human condition, thrombocytosis was not observed in these murine models and was suggested to be attributable to secondary genetic events that contribute to platelet expansion (Wernig et al. (2006)). Similar reconstitution of the TPO receptor mutation (MPL-W515L) in rodent bone marrow resulted in a myeloproliferative disease with a more rapid onset than Jak2-V617F animals that was reminiscent of primary myelofribosis including splenomegaly, hepatomegaly, and reticulin fibrosis of bone marrow (Pikman et al. (2006)). Also unlike the Jak2-V617F model, mice expressing MPL-W515L displayed dramatic thrombocytosis perhaps indicating a more dominant function of the receptor activation compared to Jak2-V617 in the expansion of this lineage. Nonetheless, these observations collectively underscore the role of both MPL-W515L and Jak2-V617F as driver mutations underlying the progression of human MPD.
A key question to the genetic basis of MPD is the role of additional genetic events that contribute to disease progression beyond Jak2 and MPL. Several lines of evidence suggest additional genetic alterations in MPD disease progression. In fact, mitotic recombination occurs frequently in MPD patients to generate two Jak2-V617F alleles indicating a selection for cell clones homozygous for the mutated kinase (Levine et al. (2005)). In this regard it will be important to develop conditional Jak2-V617F knock-in animals and to determine the phenotypic consequences of the homozygous versus heterozygous Jak2-V617F burden. Additionally, there is evidence for an inherited germline allele that precedes and predisposes patients to acquire Jak2-V617F (Goerttler et al. (2005); Levine et al. (2006)) as well as loss of chromosomal region 20q in some MPD patients. Although MPD conversion AML is observed clinically at moderate levels and activating Jak chromosomal translocations are observed in leukemia, epidemiological data suggest that it is questionable that Jak2-V617F is a genetic driver in this context suggesting additional genetic alterations are required for full leukemic transformation (Theocharides et al. (2007)). These observations notwithstanding, Jak2 inhibition with small molecule inhibitors is sufficient to modulate disease progression in pre-clinically in animal models suggesting that Jak2 activation is sufficient for maintaining MPD (Paradani et al. (2007)). It will be important to identify these additional genetic alterations and decipher how these genetic changes contribute to the disease progression of PV, ET and PMF in the context of Jak2-V617F and MPL-W515L. It will also be important to implement approaches to identify if other Jak2 pathway components are mutated in MPD patients not associated with the acquisition of Jak2-V617F, Jak2 exon12 or MPL-W515L mutations.